Finding experiments

To use incense we first have to instantiate an experiment loader that will enable us to query the database for specific runs.

targets_type iteration autoencoder_type batch_size artifacts
exp_id
42 Mnist False nomal_dim_tied 256 {'history_autoencoder': Artifact(name=history_...
43 Mnist False nomal_dim_tied 128 {'history_autoencoder': Artifact(name=history_...
44 Mnist False nomal_dim_tied 64 {'history_autoencoder': Artifact(name=history_...
45 Mnist False nomal_dim_tied 32 {'history_autoencoder': Artifact(name=history_...
46 10_Targets False nomal_dim_tied 256 {'history_autoencoder': Artifact(name=history_...
47 10_Targets False nomal_dim_tied 128 {'history_autoencoder': Artifact(name=history_...
48 10_Targets False nomal_dim_tied 64 {'history_autoencoder': Artifact(name=history_...
49 10_Targets False nomal_dim_tied 32 {'history_autoencoder': Artifact(name=history_...
targets_type iteration autoencoder_type batch_size artifacts sort
exp_id
46 10_Targets False nomal_dim_tied 256 {'history_autoencoder': Artifact(name=history_... 0
47 10_Targets False nomal_dim_tied 128 {'history_autoencoder': Artifact(name=history_... 1
48 10_Targets False nomal_dim_tied 64 {'history_autoencoder': Artifact(name=history_... 2
49 10_Targets False nomal_dim_tied 32 {'history_autoencoder': Artifact(name=history_... 3
42 Mnist False nomal_dim_tied 256 {'history_autoencoder': Artifact(name=history_... 4
43 Mnist False nomal_dim_tied 128 {'history_autoencoder': Artifact(name=history_... 5
44 Mnist False nomal_dim_tied 64 {'history_autoencoder': Artifact(name=history_... 6
45 Mnist False nomal_dim_tied 32 {'history_autoencoder': Artifact(name=history_... 7

Red best overall, and also best of subset. Bes means for accuracy max, rest min. Green best of subset.

predictions_df_0
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.9725 0.9742 0.9743 0.9775 0.9763 0.9744 0.9749 0.9748
1 0.9656 0.9711 0.9694 0.9743 0.9726 0.969 0.9703 0.9705
2 0.9615 0.9706 0.968 0.9737 0.9634 0.9603 0.958 0.9603
3 0.9577 0.9707 0.9679 0.9736 0.9486 0.9424 0.9417 0.945
4 0.9335 0.9707 0.9679 0.9735 0.9214 0.9215 0.9162 0.9194
5 0.8739 0.9707 0.9679 0.9735 0.8823 0.8894 0.8814 0.8878
6 0.8738 0.9707 0.9679 0.9735 0.8259 0.8357 0.8309 0.8459
7 0.8738 0.9707 0.9679 0.9735 0.7509 0.7709 0.782 0.7942
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.401906 0.402182 0.400775 0.404257 0.0351395 0.0356417 0.0361282 0.0350248
1 0.411027 0.408356 0.411532 0.41424 0.0506944 1.5165 0.049044 0.0497665
2 0.413155 0.40924 0.413617 0.415777 128.011 1.53466e+12 0.104961 0.0795622
3 0.413529 0.409625 0.414313 0.416035 5.74169e+12 1.73859e+24 5.99697e+11 7.6421e+10
4 0.414155 0.409752 0.414566 0.416082 2.57946e+23 inf 1.51443e+25 1.45618e+24
5 0.429525 0.409864 0.414699 0.416094 inf inf inf inf
6 0.432744 0.409999 0.414776 0.416091 inf inf inf inf
7 0.432623 0.410205 0.414882 0.416091 inf inf inf inf
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.267925 0.266109 0.266977 0.267262 0.0764037 0.0766775 0.0747766 0.072633
1 0.27 0.267454 0.269877 0.269511 0.092591 0.0982461 0.0874124 0.0866574
2 0.270812 0.267775 0.270718 0.269808 0.181491 8695.06 0.103686 0.103388
3 0.271383 0.267935 0.271334 0.269856 14576.1 9.2552e+09 10159 2789.89
4 0.272609 0.268017 0.271653 0.269869 3.08947e+09 9.85098e+15 5.10516e+10 1.21783e+10
5 0.279003 0.268097 0.271912 0.269872 6.54831e+14 1.04851e+22 2.56548e+17 5.31604e+16
6 0.279818 0.268191 0.27208 0.269871 1.38795e+20 1.11601e+28 1.28922e+24 2.32054e+23
7 0.279777 0.268331 0.27227 0.269871 2.94183e+25 inf 6.47866e+30 1.01296e+30 
predictions_df_10
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.9394 0.9535 0.9579 0.9628 0.9302 0.9356 0.9158 0.8846
1 0.9246 0.9475 0.9511 0.9585 0.9145 0.9342 0.8716 0.8591
2 0.9152 0.9456 0.9498 0.9571 0.88 0.9124 0.8265 0.8258
3 0.9098 0.9452 0.9497 0.9566 0.8421 0.8821 0.7943 0.7916
4 0.8806 0.9451 0.9495 0.9567 0.7945 0.8462 0.7569 0.7536
5 0.8323 0.9451 0.9493 0.9567 0.7376 0.7956 0.7154 0.711
6 0.8321 0.945 0.9492 0.9567 0.6773 0.7381 0.6704 0.6628
7 0.8321 0.945 0.9492 0.9567 0.6038 0.6653 0.6266 0.6101
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.396474 0.399234 0.397118 0.400997 0.192351 0.110588 194.25 323.5
1 0.41353 0.40974 0.4129 0.414831 2.5013e+09 1.7332e+10 4.86995e+15 6.0761e+15
2 0.419493 0.41211 0.415773 16318.8 1.12371e+20 1.9635e+22 1.22982e+29 1.15778e+29
3 0.421175 1.01524 0.416629 8.81034e+17 5.04826e+30 inf inf inf
4 0.422485 2.05994e+13 0.41691 inf inf inf inf inf
5 0.436508 7.20358e+26 0.417071 inf inf inf inf inf
6 0.439568 inf 0.417155 inf inf inf nan nan
7 0.439455 inf 0.41726 inf inf nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.268623 0.266593 0.267358 0.26736 0.139373 0.117621 0.314835 0.441296
1 0.272837 0.268774 0.270971 0.270213 3132.26 4635.32 999073 1.50629e+06
2 0.274721 0.269433 0.271915 1.3168 6.63914e+08 4.93512e+09 5.02064e+12 6.57535e+12
3 0.275617 0.275556 0.272554 7.6888e+06 1.4072e+14 5.2528e+15 2.523e+19 2.87026e+19
4 0.277031 35701.6 0.272898 5.65003e+13 2.98264e+19 5.59094e+21 1.26787e+26 1.25292e+26
5 0.282812 2.11124e+11 0.273168 4.15186e+20 6.32187e+24 5.95084e+27 inf inf
6 0.2836 1.24849e+18 0.273348 3.05095e+27 1.33995e+30 inf nan nan
7 0.28356 7.38298e+24 0.273539 inf inf nan nan nan 
predictions_df_20
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.8895 0.9119 0.9304 0.9357 0.8785 0.8809 0.8246 0.7873
1 0.8638 0.9004 0.9137 0.9297 0.8418 0.8677 0.7476 0.7315
2 0.8499 0.8973 0.9107 0.9275 0.7846 0.8324 0.7059 0.6908
3 0.8427 0.8969 0.9101 0.9273 0.734 0.7962 0.6689 0.6575
4 0.8138 0.8969 0.9099 0.9272 0.6795 0.7513 0.6252 0.6249
5 0.7717 0.8969 0.9098 0.9273 0.6211 0.6896 0.5839 0.5854
6 0.7716 0.8969 0.9097 0.9273 0.5573 0.6217 0.5472 0.5458
7 0.7716 0.8969 0.9097 0.9273 0.4964 0.5511 0.5143 0.5007
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.393815 0.397695 0.393258 0.398758 0.540352 12.6282 2042.09 2093.66
1 0.421537 0.415857 0.416541 8.61158e+09 1.03755e+10 1.3519e+13 5.12708e+16 3.94084e+16
2 7012.26 0.41945 0.421664 4.65015e+23 4.66121e+20 1.53155e+25 1.29475e+30 7.50917e+29
3 1.37016e+16 0.420125 0.423078 inf 2.09405e+31 inf inf inf
4 2.678e+28 0.420328 0.423595 inf inf inf inf inf
5 inf 0.420459 0.42377 inf inf inf inf inf
6 inf 0.420585 0.423853 nan inf inf nan nan
7 inf 0.420775 0.424009 nan inf nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.271418 0.268266 0.268724 0.268546 0.215259 0.212319 1.53183 2.0569
1 0.278964 0.272681 0.273929 864.386 10558.3 73355.8 7.31266e+06 8.94264e+06
2 0.948803 0.273592 0.275314 6.35269e+09 2.23792e+09 7.80861e+10 3.67481e+13 3.90366e+13
3 932442 0.273795 0.276075 4.66821e+16 4.7434e+14 8.31127e+16 1.84669e+20 1.70402e+20
4 1.30359e+12 0.273888 0.276487 3.43038e+23 1.00539e+20 8.84628e+22 9.2801e+26 7.43833e+26
5 1.82247e+18 0.27397 0.276757 2.52078e+30 2.13098e+25 9.41574e+28 inf inf
6 2.54788e+24 0.274059 0.27694 nan 4.51672e+30 inf nan nan
7 3.56204e+30 0.274189 0.277151 nan inf nan nan nan 
predictions_df_30
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.8281 0.8623 0.8954 0.8891 0.8088 0.8186 0.7288 0.6939
1 0.7974 0.8449 0.8698 0.8832 0.7589 0.7919 0.6393 0.6139
2 0.7828 0.8425 0.8665 0.882 0.6936 0.7455 0.5955 0.5662
3 0.7724 0.8418 0.8658 0.8812 0.6349 0.7016 0.5589 0.5347
4 0.7455 0.8415 0.8657 0.8812 0.5734 0.6465 0.5258 0.5053
5 0.7108 0.8414 0.8659 0.8812 0.5129 0.5884 0.4932 0.4742
6 0.7104 0.8414 0.8658 0.8812 0.4573 0.5242 0.4514 0.4398
7 0.7105 0.8414 0.8658 0.8812 0.405 0.459 0.4225 0.4086
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.392332 0.396977 0.3898 0.39654 0.939852 163.789 25688.7 25963.6
1 4109.67 0.423185 104.969 1.37296e+10 2.00048e+10 1.82068e+14 6.47113e+17 4.92275e+17
2 8.02976e+15 110.179 3.22847e+14 7.4138e+23 8.98718e+20 2.06263e+26 1.63417e+31 9.38017e+30
3 1.56943e+28 3.82779e+15 9.98935e+26 inf 4.0375e+31 inf inf inf
4 inf 1.33857e+29 inf inf inf inf inf inf
5 inf inf inf inf inf inf inf inf
6 inf inf inf nan inf inf nan nan
7 nan inf nan nan inf nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.274316 0.270822 0.270462 0.270851 0.289015 0.526565 8.64217 11.1109
1 0.793936 0.277284 0.360978 976.664 17795.6 396072 4.38253e+07 4.98728e+07
2 713950 0.359235 146412 7.17728e+09 3.77193e+09 4.2159e+11 2.20234e+14 2.17705e+14
3 9.98131e+11 486670 2.57538e+11 5.27415e+16 7.99482e+14 4.48728e+17 1.10673e+21 9.50319e+20
4 1.39543e+18 2.87796e+12 4.53013e+17 3.87565e+23 1.69455e+20 4.77614e+23 5.56162e+27 4.14831e+27
5 1.95086e+24 1.70189e+19 7.96859e+23 2.84798e+30 3.59168e+25 5.08359e+29 inf inf
6 2.72738e+30 1.00642e+26 1.40169e+30 nan 7.61276e+30 inf nan nan
7 nan inf nan nan inf nan nan nan 
predictions_df_40
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.7442 0.8027 0.842 0.8293 0.7398 0.7475 0.651 0.5991
1 0.7129 0.7811 0.816 0.8204 0.6803 0.6971 0.5404 0.4956
2 0.6968 0.7758 0.81 0.8196 0.6035 0.6439 0.4988 0.4599
3 0.6888 0.7751 0.8088 0.8193 0.5412 0.5942 0.4616 0.4346
4 0.667 0.775 0.8088 0.8193 0.4825 0.5403 0.4276 0.411
5 0.6364 0.7749 0.8085 0.8193 0.4268 0.4843 0.3953 0.3813
6 0.6364 0.7747 0.8085 0.8193 0.3764 0.4331 0.3618 0.3564
7 0.6364 0.7745 0.8085 0.8193 0.3273 0.3749 0.3398 0.3281
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.394308 0.399596 0.388105 0.398341 1.29555 2114.88 74397.2 66594.4
1 0.437422 2.05557e+09 0.42805 8.67514e+10 2.8314e+10 2.38535e+15 1.87499e+18 1.26391e+18
2 0.452615 7.18831e+22 0.43748 4.68447e+24 1.27201e+21 2.70234e+27 4.73496e+31 2.40835e+31
3 0.457902 inf 0.440222 inf 5.7145e+31 inf inf inf
4 0.460749 inf 0.440985 inf inf inf inf inf
5 0.469558 inf 0.441284 inf inf inf inf inf
6 0.471745 nan 0.441408 nan inf nan nan nan
7 0.471675 nan 0.441505 nan inf nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.278815 0.274847 0.273459 0.275008 0.353926 1.61756 19.856 20.8491
1 0.290441 356.868 0.28224 4578.13 23643.2 1.55248e+06 1.01494e+08 9.40415e+07
2 0.29404 2.10907e+09 0.284501 3.365e+10 5.01135e+09 1.65246e+12 5.10035e+14 4.1051e+14
3 0.295634 1.24721e+16 0.285462 2.47273e+17 1.06218e+15 1.75883e+18 2.56306e+21 1.79195e+21
4 0.297137 7.3754e+22 0.285961 1.81706e+24 2.25136e+20 1.87205e+24 1.28801e+28 7.82216e+27
5 0.300661 4.36146e+29 0.286263 1.33525e+31 4.77187e+25 1.99256e+30 inf inf
6 0.30125 nan 0.286468 nan 1.01142e+31 nan nan nan
7 0.301221 nan 0.286654 nan inf nan nan nan 
predictions_df_50
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.6727 0.7393 0.7898 0.7637 0.6826 0.6847 0.5667 0.5237
1 0.6395 0.7144 0.751 0.7551 0.6165 0.6183 0.4546 0.4094
2 0.6247 0.7101 0.7465 0.7542 0.5355 0.5602 0.4147 0.373
3 0.615 0.7091 0.7448 0.7535 0.4729 0.5022 0.3839 0.3522
4 0.5959 0.7091 0.7447 0.7538 0.4163 0.4438 0.3554 0.3317
5 0.5727 0.709 0.7445 0.754 0.3681 0.3906 0.3288 0.3155
6 0.5728 0.709 0.7445 0.754 0.3186 0.3431 0.2973 0.2883
7 0.5728 0.709 0.7445 0.754 0.2832 0.2905 0.2862 0.267
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.397009 0.402216 0.38788 1.75283 1.93966 7096.74 306678 174467
1 0.446041 8.81811e+09 2.59058e+08 7.16085e+13 4.72365e+10 8.01223e+15 7.73425e+18 3.31318e+18
2 0.462857 3.08368e+23 8.01559e+20 3.86677e+27 2.1221e+21 9.07697e+27 1.95315e+32 6.31319e+31
3 2.81401e+06 inf inf inf 9.53357e+31 inf inf inf
4 5.49996e+18 inf inf inf inf inf inf inf
5 1.07497e+31 inf inf inf inf inf inf inf
6 inf nan inf nan inf nan nan nan
7 inf nan nan nan inf nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.28291 0.278408 0.277079 0.289299 0.431472 4.12467 57.6568 47.0914
1 0.295947 954.831 223.894 80662.7 32253.2 4.28743e+06 2.94178e+08 2.11885e+08
2 0.299673 5.64561e+09 3.93557e+08 5.92768e+11 6.8363e+09 4.5635e+12 1.47832e+15 9.24921e+14
3 13.6885 3.33855e+16 6.92275e+14 4.35589e+18 1.44899e+15 4.85727e+18 7.42895e+21 4.03744e+21
4 1.87191e+07 1.97426e+23 1.21772e+21 3.20088e+25 3.07121e+20 5.16994e+24 3.73324e+28 1.76241e+28
5 2.61701e+13 1.16749e+30 2.142e+27 inf 6.5096e+25 5.50274e+30 inf inf
6 3.65868e+19 nan inf nan 1.37975e+31 nan nan nan
7 5.11498e+25 nan nan nan inf nan nan nan 
predictions_df_60
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.5972 0.6484 0.7067 0.6803 0.618 0.6188 0.4835 0.4652
1 0.567 0.6245 0.6715 0.669 0.5591 0.5465 0.3762 0.3439
2 0.5509 0.6193 0.664 0.6696 0.4645 0.4734 0.3379 0.3106
3 0.5437 0.6185 0.662 0.6697 0.3986 0.4095 0.3132 0.297
4 0.5286 0.618 0.6614 0.6696 0.3436 0.354 0.2875 0.2804
5 0.5065 0.6178 0.6612 0.6695 0.2996 0.3113 0.2661 0.2682
6 0.5062 0.6177 0.6612 0.6694 0.2618 0.2779 0.2432 0.2441
7 0.5062 0.6177 0.6612 0.6694 0.233 0.2331 0.2334 0.2242
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.400592 0.40445 0.389921 0.49054 3.93332 12739.7 537799 279889
1 9.71486 3.22605e+08 5.23878e+07 4.28895e+12 1.27987e+11 1.43873e+16 1.35649e+19 5.31527e+18
2 1.79304e+13 1.12814e+22 1.62094e+20 2.31598e+26 5.74982e+21 1.62991e+28 3.42557e+32 1.01281e+32
3 3.50451e+25 inf inf inf inf inf inf inf
4 inf inf inf inf inf inf inf inf
5 inf inf inf inf inf inf inf inf
6 inf nan inf nan inf nan nan nan
7 inf nan nan nan nan nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.287791 0.28281 0.28203 0.289423 0.497351 6.99165 91.9893 76.4692
1 0.326877 142.39 59.2327 31705.5 38315.7 7.40249e+06 4.6907e+08 3.43652e+08
2 34330.8 8.40639e+08 1.03744e+08 2.33014e+11 8.12128e+09 7.87913e+12 2.3572e+15 1.50011e+15
3 4.79957e+10 4.97115e+15 1.82488e+14 1.71227e+18 1.72135e+15 8.38632e+18 1.18455e+22 6.54823e+21
4 6.70999e+16 2.9397e+22 3.20999e+20 1.25825e+25 3.64849e+20 8.92617e+24 5.9527e+28 2.85842e+28
5 9.38082e+22 1.7384e+29 5.64643e+26 9.24608e+31 7.73318e+25 9.50077e+30 inf inf
6 1.31148e+29 nan inf nan 1.63909e+31 nan nan nan
7 inf nan nan nan nan nan nan nan 
predictions_df_70
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.5177 0.5779 0.625 0.5847 0.5464 0.542 0.4177 0.4029
1 0.4944 0.5506 0.5885 0.5759 0.4801 0.4581 0.3161 0.2757
2 0.4792 0.5447 0.5811 0.5767 0.3929 0.3814 0.2818 0.2455
3 0.4713 0.5434 0.5796 0.5772 0.3357 0.3237 0.2576 0.2319
4 0.4584 0.5428 0.5794 0.5774 0.2908 0.2806 0.2363 0.2189
5 0.4424 0.5428 0.5791 0.5774 0.252 0.2471 0.2179 0.2105
6 0.4422 0.5427 0.5792 0.5773 0.2182 0.2228 0.2001 0.1908
7 0.4422 0.5427 0.5792 0.5772 0.1953 0.1811 0.1935 0.1827
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.407117 1.48135 0.405271 327.569 14.5561 42173 1.18157e+06 550661
1 0.466829 3.65775e+13 2.47455e+10 1.76437e+16 5.85676e+11 4.76768e+16 2.98094e+19 1.04641e+19
2 0.486444 1.27911e+27 7.6566e+22 9.5274e+29 2.63116e+22 5.40125e+28 7.52785e+32 1.9939e+32
3 281.186 inf inf inf inf inf inf inf
4 5.47993e+14 inf inf inf inf inf inf inf
5 1.07106e+27 inf inf inf inf inf inf inf
6 inf nan inf nan inf nan nan nan
7 inf nan nan nan nan nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.293136 0.295731 0.287589 0.471978 0.636502 16.0255 168.193 116.727
1 0.309322 52151.2 2419.99 1.33588e+06 61136.3 1.71782e+07 8.56204e+08 5.23585e+08
2 0.312939 3.08409e+11 4.25722e+09 9.81671e+12 1.29582e+10 1.82842e+13 4.30265e+15 2.28555e+15
3 0.45337 1.82378e+18 7.48853e+15 7.2137e+19 2.74657e+15 1.94612e+19 2.16219e+22 9.97681e+21
4 193007 1.0785e+25 1.31725e+22 5.3009e+26 5.8215e+20 2.07139e+25 1.08656e+29 4.35505e+28
5 2.6983e+11 inf 2.31706e+28 inf 1.2339e+26 2.20473e+31 inf inf
6 3.77233e+17 nan inf nan 2.61531e+31 nan nan nan
7 5.27387e+23 nan nan nan nan nan nan nan 
predictions_df_80
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.447 0.4905 0.5387 0.5006 0.4736 0.4716 0.3644 0.3404
1 0.417 0.4656 0.4981 0.4916 0.4119 0.3856 0.2625 0.2213
2 0.4061 0.461 0.4915 0.4891 0.3276 0.3138 0.2377 0.2033
3 0.4029 0.4598 0.4902 0.4893 0.2801 0.2632 0.2163 0.1949
4 0.3934 0.4594 0.4899 0.4894 0.2429 0.2289 0.1995 0.1866
5 0.3803 0.4592 0.4898 0.4896 0.2128 0.2042 0.1852 0.18
6 0.3803 0.459 0.4897 0.4897 0.1871 0.1842 0.1677 0.1695
7 0.3802 0.4587 0.4897 0.4897 0.1709 0.1533 0.1696 0.1583
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.412624 0.428987 5.25728 206.994 67.251 67323.1 1.5699e+06 791614
1 61.9679 2.98596e+11 1.49057e+13 1.11191e+16 2.92653e+12 7.61074e+16 3.96073e+19 1.50441e+19
2 1.19777e+14 1.04419e+25 4.61203e+25 6.00421e+29 1.31475e+23 8.62211e+28 1.00021e+33 2.86661e+32
3 2.34105e+26 inf inf inf inf inf inf inf
4 inf inf inf inf inf inf inf inf
5 inf inf inf inf inf inf inf inf
6 inf nan nan nan inf nan nan nan
7 nan nan nan nan nan nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.297793 0.293884 0.311141 0.568892 0.838118 26.3605 219.248 158.456
1 0.401997 6491.14 32654.4 1.99158e+06 96557.9 2.83942e+07 1.11579e+09 7.09784e+08
2 120690 3.83872e+10 5.74402e+10 1.46351e+13 2.0466e+10 3.02223e+13 5.60717e+15 3.09834e+15
3 1.6873e+11 2.27004e+17 1.01038e+17 1.07544e+20 4.33788e+15 3.21677e+19 2.81775e+22 1.35248e+22
4 2.35891e+17 1.3424e+24 1.77728e+23 7.90279e+26 9.19438e+20 3.42384e+25 1.41599e+29 5.90381e+28
5 3.29785e+23 7.9383e+30 3.12627e+29 inf 1.9488e+26 3.64424e+31 inf inf
6 4.61052e+29 nan nan nan 4.13059e+31 nan nan nan
7 nan nan nan nan nan nan nan nan 
predictions_df_90
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.3777 0.4226 0.4579 0.4063 0.4083 0.4039 0.3001 0.2952
1 0.3566 0.3998 0.4223 0.3962 0.3457 0.3087 0.2162 0.176
2 0.3482 0.3953 0.4162 0.3946 0.2794 0.2469 0.1909 0.1643
3 0.3416 0.3932 0.4158 0.3943 0.2436 0.2139 0.1811 0.1589
4 0.3338 0.3928 0.4152 0.3946 0.2056 0.1952 0.1703 0.1559
5 0.3269 0.3926 0.4149 0.3947 0.1811 0.176 0.1617 0.1523
6 0.3264 0.3926 0.4148 0.3945 0.1619 0.163 0.1449 0.1439
7 0.3264 0.3922 0.4147 0.3946 0.1512 0.1317 0.1495 0.1395
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.434935 5.47578 12.9235 665.735 129.122 121920 2.43366e+06 1.13164e+06
1 3.09453e+10 1.75089e+14 3.85059e+13 3.58653e+16 5.66694e+12 1.37862e+17 6.14055e+19 2.15093e+19
2 6.0483e+22 6.12285e+27 1.19143e+26 1.93668e+30 2.54588e+23 1.56182e+29 1.55069e+33 4.09854e+32
3 inf inf inf inf inf inf inf inf
4 inf inf inf inf inf inf inf inf
5 inf inf inf inf inf inf inf inf
6 inf nan nan nan inf nan nan nan
7 nan nan nan nan nan nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.302113 0.31301 0.335783 0.784144 1.15231 44.0107 312.26 214.05
1 1401.61 108044 68959.3 3.53244e+06 156842 4.75182e+07 1.58767e+09 9.57954e+08
2 1.95917e+09 6.38934e+11 1.213e+11 2.59581e+13 3.32435e+10 5.05775e+13 7.97846e+15 4.18165e+15
3 2.739e+15 3.77836e+18 2.1337e+17 1.9075e+20 7.04613e+15 5.38333e+19 4.00939e+22 1.82536e+22
4 3.82922e+21 2.23434e+25 3.75321e+23 1.40171e+27 1.49347e+21 5.72987e+25 2.01482e+29 7.96803e+28
5 5.35341e+27 inf 6.60197e+29 inf 3.16548e+26 6.09871e+31 inf inf
6 inf nan nan nan 6.70941e+31 nan nan nan
7 nan nan nan nan nan nan nan nan 
predictions_df_100
Accuracy over iterations evaluations_feature_classifier
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.3121 0.3456 0.3711 0.3208 0.342 0.3365 0.2589 0.251
1 0.2964 0.3299 0.3434 0.3083 0.2858 0.2578 0.1847 0.1554
2 0.2886 0.3236 0.3375 0.3077 0.2305 0.2071 0.1673 0.1415
3 0.2833 0.3225 0.3372 0.3086 0.2034 0.1851 0.1526 0.1391
4 0.2787 0.322 0.3368 0.3085 0.1747 0.1721 0.1449 0.1377
5 0.2724 0.3218 0.3368 0.3083 0.1512 0.1594 0.1369 0.1346
6 0.2723 0.3217 0.3365 0.3077 0.1423 0.1478 0.1268 0.1196
7 0.2723 0.3214 0.3365 0.3077 0.1343 0.1141 0.1275 0.1195
Loss over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 34.4267 187.285 22.7173 11225.7 550.669 244099 4.03741e+06 1.88882e+06
1 6.61017e+13 6.51473e+15 6.84727e+13 6.05835e+17 2.44699e+13 2.76091e+17 1.01882e+20 3.5912e+19
2 1.29197e+26 2.27819e+29 2.11864e+26 3.27143e+31 1.09931e+24 3.1278e+29 2.57286e+33 6.84293e+32
3 inf inf inf inf inf inf inf inf
4 inf inf inf inf inf inf inf inf
5 inf inf inf inf inf inf inf inf
6 nan nan nan nan inf nan nan nan
7 nan nan nan nan nan nan nan nan
MAE over iterations autoencoder
normal_dim_iteration256 10_Targets normal_dim_iteration128 10_Targets normal_dim_iteration64 10_Targets normal_dim_iteration32 10_Targets normal_dim_iteration256 Mnist normal_dim_iteration128 Mnist normal_dim_iteration64 Mnist normal_dim_iteration32 Mnist
0 0.389266 0.519096 0.381135 2.91465 2.28748 78.883 468.678 304.72
1 118716 1.31305e+06 137995 1.92051e+07 395787 8.51905e+07 2.37958e+09 1.36068e+09
2 1.65971e+11 7.76489e+12 2.42741e+11 1.41128e+14 8.38893e+10 9.06751e+13 1.1958e+16 5.93964e+15
3 2.32034e+17 4.59179e+19 4.26986e+17 1.03706e+21 1.77808e+16 9.6512e+19 6.0092e+22 2.59276e+22
4 3.24392e+23 2.71537e+26 7.51075e+23 7.62072e+27 3.76873e+21 1.02725e+26 3.01978e+29 1.13178e+29
5 4.53513e+29 inf 1.32116e+30 inf 7.98803e+26 1.09337e+32 inf inf
6 nan nan nan nan 1.69311e+32 nan nan nan
7 nan nan nan nan nan nan nan nan 
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)

/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)

/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)

/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)

/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)

/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)

/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:397: UserWarning: Warning: converting a masked element to nan.
  dv = (np.float64(self.norm.vmax) -
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:398: UserWarning: Warning: converting a masked element to nan.
  np.float64(self.norm.vmin))
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:405: UserWarning: Warning: converting a masked element to nan.
  a_min = np.float64(newmin)
/home/hicky/anaconda3/lib/python3.7/site-packages/matplotlib/image.py:410: UserWarning: Warning: converting a masked element to nan.
  a_max = np.float64(newmax)
<string>:6: UserWarning: Warning: converting a masked element to nan.
/home/hicky/anaconda3/lib/python3.7/site-packages/numpy/ma/core.py:711: UserWarning: Warning: converting a masked element to nan.
  data = np.array(a, copy=False, subok=subok)